The behaviour of a quadrupole mass spectrometer (QMS) for variation in rod size has been investigated using a numerical approach reported previously. Behaviour is found to vary significantly with the change in rod size and the optimum rod size suggested is closer to that used in some commercial QMS instruments than to the value from analytical approaches. The results allow an upper limit to manufacturing tolerances to be inferred. The present work also confirms that the size of the QMS housing has only a very small effect on QMS behaviour. Copyright # 2001 John Wiley & Sons, Ltd.Quadrupole mass spectrometer (QMS) systems are usually manufactured with circular section electrodes although the original design assumes the use of hyperbolic section electrodes. We previously reported 1 predictions of performance when circular cross section electrodes are used instead of hyperbolic section ones. The results showed differences in predicted QMS behaviour for the two geometries. In particular, the circular section system had lower resolution, mass peaks had a long low mass tail, and the ions had to experience more cycles of the radiofrequency (rf) field for good filter performance to be obtained (longer system).Further application of the model has allowed some design features of QMS systems to be investigated. In particular, the effects of rod size, the housing, and a few features of the low mass tail have been examined.
ROD SIZEWhen hyperbolic cross section electrodes are used the surface cross section should be located on a right-angled hyperbola with its axis at the QMS axis. For a perfect system the only effect of changing electrode size is to require changed electrode separation and different applied voltages; QMS behaviour is unaffected by electrode size if the ion source is scaled such that the source radius is scaled so that the ratio of radius to rod separation remains the same. If the ion source is not scaled then transmission increases as the rod size increases because ions close to the axis have a higher transmission probability than those distant from it. When circular cross section electrodes are used they are positioned so that the electric field approximates to that for hyperbolic electrodes. The field produced depends on the ratio of the rod radius, r, to the distance from the rod surface to the QMS axis, r 0 . Changes in the ratio r/r 0 affect QMS performance.Early theoretical considerations of QMS design 2,3 determined an optimum value for r in terms of r 0 . The earliest such work was an empirical treatment 2 which suggested that a value of r = 1.148 Â r 0 should be used. This result was incorrectly quoted by Paul et al. 4 as r = 1.16 Â r 0 ; this was subsequently quoted elsewhere and adopted by some QMS manufacturers. A more formal treatment by Denison 3 expanded the field of the circular electrode quadrupole as an infinite series of multipole components with terms present only for 4-poles (quadrupole), 12-poles, 20-poles, and so on. The field was determined numerically and compared to a multipole expa...